1. Pay for Performance Conference Jon Kingsdale, Ph.D. John Freedman, M.D., M.B.A.

2. Outline u Approach to, and rationale for “value-based” tiering u Collaboration with providers to develop “value- based” metrics u Member response to tiering

3. Decrease Medical Trend & Improve Quality & Service Supply Side u Risk Contracting u P4P u Selective Contracting u Profiling u UR\PA u TIERING Demand Side u Benefits u Cost-Sharing u HRA\HSA u Disease Management u Health Promotion u TIERING Network Tiering bridges the boundary between supply- side and demand-side initiatives

4. Plan Design Overview u PPO Benefits u Phased, multi-year, approach beginning 7/1/04 * –In-network providers covered at different levels based on quality and efficiency measures –Out-of-network covered at 80% after deductible u Efficiency and quality measures –Began with index scores for hospitals –3 hospital inpatient specialties –Add PCPs and specialists in future u Variable co-pay based on provider selection u Core medical & Rx management * State’s open enrollment effective 7/1/04

5. Year 1: Fiscal Year 2005 Model (7/1/04 – 6/30/05) Example of Hospital Index Year 1: FY 2005 Better Quality Good Efficiency Standard co-pay Better Quality Better Efficiency Lower co-pay Good Quality Good Efficiency Higher co-pay Good Quality Better Efficiency Standard co-pay Hospital Index (Inpatient) Efficiency Quality

6. Year 1: Fiscal Year 2005 Model (7/1/04 – 6/30/05) Actual Hospital Index (Inpatient) Cost Efficiency Quality Lower scores Higher score Higher Quality/ Good Efficiency Standard Copay (50% of hospitals) Good Quality/ Higher Efficiency Lowest Copay (25% of hospitals) Higher Quality/ Higher Efficiency Highest Copay (25% of hospitals) Good Quality/ Good Efficiency

7. Hospital Cost & Quality Measures u Cost –Adjusted average cost per case : Contracted rates Average length of stay Service mix –Case-mix and severity adjusted u Quality –Adjusted mortality rate –Adjusted complications rate (AHRQ) –NHVRI/JCAHO measures –Leapfrog (CPOE, ICU Staffing, Safe Practices) –Volume –Credentialing status

9. Hospital Response u “Right product, right concept” u Upset by initial lack of consultation u Methodology stinks

10. Refinements via collaboration u Feedback on hospital inpatient metrics u Extensive network involvement –Network hospitals individually and collaboratively –Expert Panel convened throughout summer, 2004 –Invited Hospital Association to have leading role u Great respect for process and grudging acceptance of outcome u One tier-3 hospital given consulting assistance & pulled itself up to tier-1

11. Better Quality Good Efficiency Standard co-pay Better Quality Better Efficiency Lower co-pay Good Quality Good Efficiency Higher co-pay Good Quality Better Efficiency Standard co-pay Physician Index (Outpatient & Inpatient) Efficiency Quality Original 3 Year Proposal: PCP’s: FY 2006 Specialists: FY 2007

12. Provider Education & Outreach 2.0 u PCP ratings development began July, 2005 u Began discussion with Central Physicians Committee in Sept. 2004 –Review industry trends and Tufts HP strategy related to quality and efficiency measurement –Overview of plan design and tiering methodology by Ms. Mitchell u Reached out to Massachusetts Medical Society u Physician Quality Measurement Expert Advisory Panel empowered to help define quality and efficiency metrics in conjunction with Central Physicians Committee u Value-based ratings using cost (episodes of care) and quality (HEDIS & patient satisfaction)

13. 1.Sensitize beneficiaries to value [quality & price] 2.Enable shopping (“transparency”) –3-tier Rx –Value-scoring providers –Decision-support tools 3.Align contracting strategy (P4P) How to Design Products and Deploy Information to Improve Value:

14. Sensitize Members to Value in Plan Design HospitalPediatricsObstetricsAdult Med/Surg Hospital A$200 Hospital B$400$600$400 Hospital CN/A$400$600 Inpatient Copayment by Value Tier

15. Sample Web Screen Enables Shopping

17. Percentage of Cases at Tier 1 Hospitals Among Persisting Members (Baseline vs Year 1)

18. Percentage of Cases at Tier 1 Hospitals for Termed vs New Members

19. Health Plan Decision Making: Factors Considered - Major Categories Multiple responses allowed. Sample size: 395

20. Health Plan Decision Making: Factors Considered - Details u “Premium cost” was the most frequently considered factor by new members. Out-of-pocket costs was the least frequently mentioned reason Multiple answers allowed. Sample sizes: New=203, Renewed=203

21. Health Plan Decision Making: The Reasons that Put Navigator Ahead u Those new members who also seriously considered plans other than Navigator decided on Navigator, because it provided freedom to choose a doctor and their doctors/hospitals were in the network. Again, OOP was least consideration. Multiple answers allowed. Sample sizes: New=109, Renewed=64 (Asked only to those who considered other health plans.)

22. Information Sources: Tufts HP Web site – Info. Sought u Two-thirds of those who visited Tufts HP’s Web site (30% of members) looked up providers. Information about Tufts HP, in general, was also sought by about a third of them. u Fewer people looked for information about drug tiers/copays, hospital copay levels, and the hospital quality profile. Sample size (THP Web site visitors): 113

23. Experiences of Renewed Members: Usage u Of those members who reported that they or their family members had been admitted to a hospital while being covered by the Navigator plan, only 9% said that they used the online tools to find information about the hospital before the hospitalization. Sample sizes: Admitted to a hospital=203, Used online tool=66

24. Experiences of Renewed Members: Satisfaction u 89% of renewed members completely/very satisfied with the Navigator plan u 77% of renewed members completely/very satisfied in 2005 CAHPS survey u Satisfaction score of those Navigator members who were admitted was slightly lower than for members without such an experience. This finding is consistent with results from other studies, which find that healthier members tend to be more satisfied. Sample sizes: Overall=203, Hospital-Yes=66, No=137, Online: Yes=6, No=60 Completely/very/somewhat satisfied = 96.6%

25. Summary u Because of direct influence on providers and the providers’ influence on members, credibility of metrics is crucial u Collaboration with providers to develop “value-based” metrics is key process step u Provider response has been great respect for process and grudging acceptance of metrics & product u Early member response to metrics & copay tiering is marginal, but change on the margin may suffice